Phillip Cloud
Phillip Cloud
Also, we're not going to deprecate union. Vertical concatenation is one use for union, but it's also a part of a suite of set operations that have nothing to do...
> Thanks Philip, I can agree with all that. I can shut up now ;) I very much appreciate your continued engagement with the project :)
@jpohanka Thanks for opening the issue! Would you mind elaborating a bit on what you're using `parse_sql` for? Would love to hear about use cases in the wild.
Roger that! It's definitely got a lot of potential.
One option if you must have the flags before we investigate here is to define a [builtin UDF](https://ibis-project.org/reference/scalar-udfs#examples). Something like this: ```python @ibis.udf.scalar.builtin(name="regexp_search") def re_search(s, pattern, flags): ... ```
I am hitting this now too, when building a nix closure under qemu (aarch64) on an x86_64 machine, that itself is a cloud VM, so there are at least two...
Thanks!
I think that makes sense. Aligning with the "projected" form of this construction for consistency seems like a good thing to have. The tricky bits here are going to be...
The current use case of ```python t.group_by(g).order_by(o) ``` is automatic windowization (broadcasting in numpy lingo) of any aggregates in a subsequent `mutate` call. ```python t.group_by(g).order_by(o).mutate(t.x.sum()) ``` desugars to ```python t.mutate(t.x.sum().over(group_by=g,...
IMO it seems odd to have two nearly identical constructions, one that has well-defined behavior, and one that errors even though it could have well-defined behavior.